Yes, in the sense that fusion requires incredibly accurate magnetic fields to maintain, and the second there's any issue with the reactor chamber, the reaction will just stop. The reactor itself cannot explode in any way shape or form, because there is nothing in there to explode. It also doesn't produce any radioactive isotopes while running, it just fuses(hence fusion) hydrogen into helium, just like the sun does. You can just capture this helium and sell it to make baloons if you want.

The reaction itself kicks off a huge amount of neutron radiation, which eventually makes the reactor chamber radioactive - that is the only radioactive waste that will have to be disposed safely eventually. But neither the fuel nor the resulting product are radioactive.

General Fusion's approach doesn't use magnetic confinement. Instead, they use liquid metal and pistons to create the pressures needed for fusion. The liquid metal then absorbs the heat energy which is extracted in the usual way.

It does use magnetic confinement. That central column is part of it. Then it adds the compression.

What is the sort of lifetime (ball-park) that one might expect before neutron saturation of the reactor walls becomes a serious concern and the reactor has to be scrapped?

You can read about stuff like that in some of the ITER technical reports. They actually want to use that neutron radiation to generate tritium, and feed that back into the reactor.

I don’t think the reactor would be scrapped, just shutdown for maintenance.

I imagine you just replace the shielding on the inner walls, not the entire reactor.

The General Fusion design covers the walls with a molten mix of lead and lithium held in place by spinning the chamber. The lithium turns to tritium when you bang on it with a neutron to crease more inputs to the D-T fusion process. And lead doesn't really mind getting hit with neutrons. From the diagrams on Wikipedia[1] it looks like there is some normal metal exposed which might get neutron activated.

[1]https://en.wikipedia.org/wiki/General_Fusion

The liner is liquid metal

Most of the energy that is produced is in the neutrons, so it will be transferred as heat in whatever shield captures the neutrons and which will become radioactive.

So most of the heat will have to be extracted from a radioactive material, with similar precautions like in fission reactors, where the heat is extracted from the radioactive nuclear fuel.

I am very skeptical that fusion of deuterium with tritium or of deuterium with deuterium will ever produce "clean energy", even if they are the easiest fusion reactions, due to the relatively low temperatures required for them.

It still remains to be proven whether the radioactive waste for a fusion reactor of the kinds attempted now will be less than for a fission reactor.

>>It still remains to be proven whether the radioactive waste for a fusion reactor of the kinds attempted now will be less than for a fission reactor.

I wonder, how can this possibly be even a question? Fission based reactors obviously have the same or worse problem of irradiating the entire reactor enclosure and everything around it, so that's at best the same as a fusion reactor + they produce tonnes of very highly radioactive waste that will be radioactive for millennia.

Materials activated through neutron bombardment aren't radioactive for anywhere near as long. And to add to that, nearly all elements produced in a fission reactor are highly toxic in addition to being radioactive - in a fusion reactor if your steel containment chamber becomes activated, you just have radioactive steel, not one of the many many dangerous heavy metals produced through fission.

What you say is true of light water reactors, a 1950s design.

LFTR would be far better. Neutrons breed new fuel, and it consumes virtually all the radioactive fuel with "waste" that rapidly becomes non-dangerous within a month.

So I would guess neutron degradation of the equipment/vessel/reactor will probably be a similar problem in both cases.

The waste really only remains radioactive for a month or so? Not 10k years? Is that just because the other radioactive waste is consumed as fuel, or is that generally true of what it produces?

I'm not a nuclear engineer, but one of the LFTR videos seemed to comprehensively break down the nucleotide chains, and yes the fundamental breeding ability I think helps.

IIRC people have stated that the breeding could also reprocess the bad millenial-scale waste from LWR into usable fuel or other isotopes in MSRs, although maybe not in LFTR.

If we had a decent LFTR industry, we could probably have more specialized MSRs to deal with the "legacy" waste and breeder reprocess into new fuel or other stuff.

The chemists from ORNL seemed pretty amazing. They had all the work done for separating out the byproducts for use in other applications.

Fission is a chain reaction, Fusion is not. A fusion reaction is so unstable that if the reactor failed the conditions for sustaining fusion would immediately stop.

It's basically the hottest plasma we can make suspended in a magnetic donut surrounding by near absolute zero temperatures.

https://en.m.wikipedia.org/wiki/Tokamak

Probably a dumb question, but what are the potential environmental impacts of manufacturing lots of helium? Is it a greenhouse gas? Even if we pivoted to 100% fusion in the next 100 years, is there a chance of releasing enough helium that it would be a problem in practice? I assume we won't need to worry about running out of hydrogen considering all of the water in the world (provided we can efficiently get hydrogen from the more abundant ocean water rather than the limited fresh water)?

> Is it a greenhouse gas?

Certainly not in the conventional sense (although perhaps there's another way that helium might act similarly to a GHG that I'm not aware of).

Greenhouse gasses are molecules, i.e. multiple atoms bonded together. Those molecules can absorb photons of infrared light, which cause them to vibrate (as if the atoms were held together by springs). After some time, the vibration stops and an infrared photon is emitted.

The problem is: those photons are emitted in a random direction, unrelated to the photon that was absorbed. Half the time they will go roughly upwards, half the time they'll go downwards.

A photon of visible light (from the Sun) can travel down through the atmosphere without interacting much with the greenhouse gasses, since it has too much energy to be absorbed. This visible photon can be absorbed by other materials at ground level, e.g. by a plant, and its energy will eventually result in around 20 lower-energy infrared photons being emitted back up (on average).

These 20 infrared photons are readily absorbed by the greenhouse gasses, and each time they're absorbed, they get re-emitted in a random direction: half the time heading upwards again, but half the time heading back to the ground. This is how energy gets "trapped" by greenhouse gasses.

Helium is almost completely unreactive: it doesn't form molecules in the atmosphere, it just bounces around as individual atoms. Without bonds to vibrate (or asymmetries to spin), the only way it can absorb energy is by speeding up, and even this isn't very effective since its mass is so low. Fast-moving helium is also more likely to escape the Earth's gravity completely.

Hmm. I'd imagine all matter than interacts with photons to be GHGs in some format.

The frequency band for absorption and emission might be different, but the helium could absorb a shorter wavelength, then release a longer wavelength (not infrared) that can be absorbed by something else, then emitted yet again to be infrared

In fact, my understanding is that, not counting nuclear fusion, helium is a non-renewable resource critical for some kinds of critical uses like medicine; and that our current habit of putting it in balloons is considered rather reckless by people taking a longer view of things. If we could capture that helium, it would make things a lot better.

Exactly, helium is likely bound to exhaustion along with natural gases deposits: the only source of helium, currently. It can't be replaced with another gas, at least not in supercoducting magnets, so it's a problem. In the future, fusion power plants will be optimised to reduce helium losses, recycle helium produced by the reactions and may also have to extract helium from the atmosphere. The latter is a rather expensive process given that the concentration is just a few ppm.

Capturing and using gases doesn't typically prevent them from ending up in the atmosphere, unfortunately.

Like others have said though - helium escapes our atmosphere naturally and quite literally leaves Earth. That's how we're loosing all of our supply - it's just being vented into space constantly.

Helium doesn't persist in the atmosphere. The mean free velocity of helium gas is above escape velocity, so all of it leaves Earth for outer space.

Whoa, I didn't realize that was possible. Fascinating.

Don't worry, even if all energy humans used was from D+T fusion today it would only create 60% of the helium that we use. Also, helium is so light that it leaves the atmosphere so we really don't have access to much anyway.

What do we use it for? Isn’t it non reactive?

It's most valuable use is as a cryogenic. If you want sub 10K there are very few materials to choose from.

The lower the temperature, the higher your critical current is for a given superconductor. The reason HTS are a big deal isn't that you can have a superconductor at liquid nitrogen temperatures, but that you can have a very high field at liquid helium temperatures.

Rough estimate, meeting humanity's energy demand (132 PWh per year - that's all energy, not just electricity), with a reasonably efficient fusion station, we'd make 3,000-5,000 tonnes of "waste" helium per year. But as others have alluded to, it's quite a useful element and the general concern is we don't have enough.

Not much - it's lighter than the atmosphere so rises to the top and floats off into space.

UK accents will acquire a slightly higher pitch over time

Yet another way Monty Python was ahead of its time.

We have a shortage of helium on Earth, we are running out fast, all hydrogen produced can be put to good use.

That does not seem right, we have literal oceans of hydrogen?

sorry I meant helium, that was a brainfart

Yes but no at the same time.

We still have plenty of helium. It's a byproduct of oil extraction and is still often vented because it would not be profitable of capturing it.

Helium balloons will stop working and everyone will have high- pitched voices.

I had assumed both were chain reactions due to stars. That’s interesting

Well, technically, that's right. Fusion is a chain reaction. (albeit, "chain" might be a stretch here. Let's say it is a continuous process in stars.) But it's driven by gravity, not neutrons. You just need to bring about 70 times the mass of Jupiter into one place and that's it. Child's play, essentially ;).

The enormous pressure inside stars is what sustains the reaction. Neutrons coming off a fusion reaction, even if moderated, can’t really “squeeze together” other particles in any sense.

Also, per unit volume, the sun produces about as much power as a compost pile.

This lead me to the following - what if you make a compost pile the size (mass) of Sun? Meaning, it won't be made of hydrogen, but rather some carbon-based molecules. I'm not sure about other atoms in these molecules, but I think carbon is stable enough not to initiate nuclear reactions. So probably fusion won't start. What then?

Gravity would compress it. Depending on the exact composition fusion could start, ie it would become a star. According to Wikipedia carbon fusion requires a mass of 8 suns to start but there's a bunch of hydrogen in compost so maybe that would fuse.

Carbon is capable of fusion, in a sufficiently compressed stellar mass. Everything up to Iron-56 can theoretically sustain fusion.

Maybe it could work via the CNO cycle: https://en.wikipedia.org/wiki/CNO_cycle ?

Does the volume of the sun include its Corona in that last statement?

No, iirc that statement is for the part of the sun where fusion actually happens. Per volume, not a lot fusions happen in the plasma on the sun. The crux is that the volume of the compost heap grows order cubed to the radius, while the area through which the heat escapes grows quadratic. And the sun is like, a _really_ big compost heap.

In fusion on earth, we want to be considerably more efficient than the fusion process in the sun, as we don't have as much space to work with. ITER is already a pretty big machine.

That’s just the core. Go out to the photosphere, let alone the corona, and it’s about about a thousand times less, 0.27 W/m^3: https://www.wolframalpha.com/input/?i=%28solar+luminosity+%2...

Fusion can be a chain ration. In stars its caused by intense uninterpreted gravitational pressure. In the H-Bomb its triggered by a smaller fission explosion. Fusion chain reactions just require so much energy its hard (if not impossible) to harness on a commercial level. One small break in the the chain reaction containment and it falls apart.

Which is infinitely better then rapidly overheating and vaporizing and melting the reactor housing and spreading radiation which will stay dangerous for 200 years.

A good way to think of the difference is:

Fission requires effort to control. When things go wrong it isn't controlled.

Fusion requires effort to create. When things go wrong it isn't created.

LFTR fission reactions are self-regulating in many ways due to the liquid fuel and thermal expansion abilities, and even if something goes wrong, there is a melt plug that will automatically stop a runaway reaction.

That seems like a faulty explanation. Uncontained fusion is massively energetic and destructive. The device in this article is analogous to an H-bomb weapon with a pneumatic rather than fissile primary.

Fusion bombs are when you put fusible materials next to fission explosions.

Yes, they are massively energetic, (we wouldn't be harvesting power from it if they weren't.) However, they require a very high input energy to trigger the release of the output energy.

With a fission reactor meltdown, the way you get there is by pulling out dampening rods or boiling off all the water, but otherwise leaving the fission rods in the same place.

With a fusion reaction, you have to be constantly providing both the energy to keep the fusion going AND the input material to be fused. Interrupt one or the other and fusion stops.

I know it seems weird that the bigger energy release is safer, but that's how it is. It's the difference between requiring constant input into the system to produce power vs an idle system with no input producing power.

Their explanation is not really faulty. FYI hydrogen bombs are mainly destructive due to fission, not fusion. The fusion step is primarily for bombarding the dangerous fissile fuel with neutrons.

That a fusion bomb has significant energy from fission doesn't seem germane. It still has a great deal of energy from fusion alone.

I'm not trying to scaremonger fusion energy, but I think it's intellectually dishonest to portray is as fundamentally sound, with a binary outcome of either inertia or safe energy. This design relies on spherical compression to both initiate and confine the fusion. We should not discount the possibility that if it instead creates a cylindrical or elliptical confinement due to malfunction, it will just explode, at a minimum destroying the device. We know it is possible to initiate fusion with radial compression in a cylinder, because that's how an H-bomb secondary works.

The main safety factor in these things comes from the fact that a fusion weapon needs hundreds of kilos of hydrogen, and they are experimenting with much smaller masses. That limits the destructive potential.

> but I think it's intellectually dishonest to portray is as fundamentally sound

No offense but I think this sentiment is a huge problem with Hacker News. I have a minimal understanding of fusion energy production from an engineering perspective but I know more than the vast majority of the world due to my background.

You are smart but not a subject matter expert in this topic. Fusion is not a sustainable a reaction while fission is. Fission happens on its own, fusion requires something else to happen first (on Earth). Please don't insult the people in this thread who are subject matter experts in the field by implying they are intellectually dishonest. I am not a subject matter expert but recognize that a few of them are commenting in this thread with details that I have learned from other experts.

It's not dishonest to portray it as fundamentally sound, because it is. Your argument is that 'If scaled up several orders of magnitude this device could cause dangerous explosions.' This ignores two very important realities:

1) There isn't a reality in which these devices get scaled up to that size. 2) The real danger with fission is radiation, not explosions, which fusion reactors will produce in smaller quantities than a banana farm.

Pure fusion power, even in its largest, most powerful, Elon Musk fever-dream incarnation, is safer than even the safest fission reactor, because there is no way for it to create a boom larger than it's vessel was designed to produce.

It is more intellectually dishonest to bemoan the danger of "uncontained fusion" by citing the hydrogen bomb.

Sort of, but it's a vast oversimplification. Fission is very safe relative to something like coal because all of the pollution is concentrated in barrels instead of pushed into the atmosphere, but has the risk of meltdown. To be fair lots of other non-nuclear plants have had disasters that released toxic gasses around the globe too.

Fusion in theory has no pollution at all, but that's theoretical until large scale fusion plants are built.

Fusion reactors don't create fission products, obviously, but why wouldn't fusion create radioactive waste via neutron activation? Materials exposed to neutron flux, in any kind of device, may be activated into radioactive isotopes.

Yes, but first wall materials are chosen to have a high melting point and short half life when activated. Close the building off for 100 years then scrap it. It's a far cry from the myriad of nightmare scenarios fission plants need active control against.

Closing the building off for 100 years might work fine, but that can't be the solution to every problem, it wouldn't be economically viable. Ostensibly simple matters like routine maintenance are very complicated propositions for fusion reactors; you can turn the reactor off but it will still be too radioactive for anybody to work inside. So you either need some sophisticated robotics to repair anything that might ever need repairing, or you have to consider the entire reactor to be disposable.

Of course, repairing things inside a fission reactor is no less nasty, but fission reactors are comparably much simpler and much smaller. Swapping a fission reactor out with a new one is comparably much easier than with a fusion reactor.

For General Fusion, about the only part exposed would be the central column, which I suspect they'll design to be easily replaced; pull out from the top and drop in a new one. Aside from that, the burning plasma is just surrounded by liquid lead/lithium.

MIT's Commonwealth uses an inner wall that's 3D-printed and designed to be replaced annually. They've tested joints in their superconducting tape, which will let them open up the coils on hinges so they can drop in a new inner wall. They'll surround that with liquid beryllium/lithium as coolant and breeding blanket.

Another byproduct of fission is "long-time nuclear waste warning booty shorts" https://twitter.com/mochasucculent/status/125638615465338060...

Basically yes. Both fission reactors and fusion reactors generate a bunch of neutrons.

With a fission reactor you have to expose uranium to those neutrons and when it gets hit it either splits into dangerously radioactive components or absorbs a neutron and becomes dangerously radioactive plutonium. There are pipes and stuff that become a bit radioactive too but that's not the part that people are afraid of when an accident happens. Also, the dangerously radioactive byproducts continue to release a lot of energy after the chain reaction is shut down, about 10% of the power the reactor was run at. So even after a fission reactor is shut down you need to keep cooling it otherwise things melt and people become unhappy.

With fusion you do need to keep exposing lithium to the neutrons from the reaction to make more tritium fuel. And tritium is radioactive. But if it escapes it'll go straight up into the upper atmosphere where it won't particularly bother people and then dilute. You've still got the radioactive pipes problem you do with fission but, again, that's not really the part that people are afraid of. And once you stop containing the plasma energy generation stops instantly.

I'm generally inclined to say that the risks involved in fission are worth the benefits but those risks are worth taking seriously and require careful government regulation more so than other forms of power. With fusion, on the other hand, I'd a lot less concerned with someone building a power plant up wind of me than a coal plant even with similar levels of regulation. Just as long as they aren't deliberately exposing uranium to the neutrons or something like that.

Fusion byproducts are by and large significantly safer than fission.

Fission byproducts are well known to just be a ton of heavy metal junk that decays over thousands of years.

I'm sure there's fusion byproducts that are nasty too, but fundamentally you can't get byproducts as high up on the elemental chart as you can with fission strictly because fusion is merging 2 smaller atoms into 1 larger atom. With fission you start with a larger atom and break it into 2 smaller ones. Oversimplified, but you get the idea.

This is the 1990's view on the subject. There were several projects to address the fission leftover problem. Some of these:

- Method to Reduce Long-lived Fission Products by Nuclear Transmutations with Fast Spectrum Reactors https://www.nature.com/articles/s41598-017-14319-7

- Fast-neutron reactor https://en.wikipedia.org/wiki/Fast-neutron_reactor

- Evolution of transuranium isotopic composition in power reactors and innovative nuclear systems for transmutation https://inspirehep.net/literature/1243003

And few other things. It is possible to have nuclear power with shorter byproducts than thousands of years.

Given that the US has only 1 reactor that was commercially put into operation after 1990s [1], it's a pretty realistic way to look at it in the US.

Newer reactors and spent fuel reprocessing are definitely ways to solve the issue, but the fact is you still create those nasty byproducts in the traditional reactors today.

[1]: https://en.wikipedia.org/wiki/Nuclear_power_in_the_United_St...

Well if you're talking about a in-development fusion reactor, then it's fair to use in-development fission reactors for comparison.